Cloudflare Stream thumbnail fit modes vs Sume Timeline fit modes
Cloudflare Stream fit: crop, clip, scale, fill. Sume Timeline 1.0 fit: cover, contain, stretch, blur. Which names map and where they differ.
Cloudflare Stream's fit has four values (crop, clip, scale, fill), and Sume's Timeline 1.0 fit has four too (cover, contain, stretch, blur). Crop is roughly cover, clip is roughly contain, and scale is stretch. Cloudflare's fill pads with black, and Sume's blur fills the space with a blurred frame, which is the one with no Cloudflare equivalent.
The Cloudflare definitions come from its Display thumbnails page and Sume's from the Timeline 1.0 doc, both read on 2026-10-03.
What do Cloudflare's four fit values mean?
The default is crop and the default box is 640 by 640. Cloudflare defines the values as follows: crop cuts parts of the video that do not fit the given size, clip preserves the entire frame and decreases the size of the image within the given size, scale distorts the image to fit, and fill preserves the entire frame and fills the rest of the requested size with a black background.
What do Sume's fit values mean?
Each video[] slot in a Timeline 1.0 render takes fit: cover (the default), contain, stretch or blur. The default output is 1080 by 1920, and output.width and output.height accept even integers from 256 to 2160. A 16:9 clip in a vertical frame is the standard case where the choice matters.
Timeline compose, which puts one still and one video on screen at once, takes cover, contain or stretch for image_fit and video_fit. blur is not a compose fit.
| Behaviour | Cloudflare Stream | Sume Timeline 1.0 |
|---|---|---|
| Fill the box, cut the overflow | crop (default) | cover (default) |
| Whole frame inside the box | clip | contain |
| Distort to exact size | scale | stretch |
| Whole frame, bars filled | fill (black) | blur (blurred fill) |
Where do the names mislead?
Two places. First, Cloudflare's clip and Sume's contain both keep the whole frame, but the Cloudflare page says a clip result is smaller than the requested box, while a Sume slot is always the output size, so what fills the leftover space is your call (and blur is the built-in answer).
Second, neither crop nor cover follows the subject: both cut from the centre. For an off-centre crop on Sume, use a crop in video filter first. Cloudflare's page documents no crop gravity for thumbnails.
How do you preview a fit choice cheaply?
Run the plan first. POST /v1/timeline-1.0/plan runs the schema and pure compiler without creating a job or reserving credits, and returns duration_seconds, segment_count and an estimated cost. It does not render pixels, so it will not show you the framing.
To see the framing, render a short test with a few seconds of audio and pull stills from the result with video frames. A 1-minute render costs $0.10, which is cheap next to publishing the wrong crop.
Which fit should a vertical ad use?
For a 16:9 source in a 1080 by 1920 frame, cover fills the screen and loses the sides, contain keeps everything in a thin horizontal band, and blur keeps everything and fills the rest with the clip itself, which is usually the safest for product shots. Use stretch only when the aspect ratios already match. If you build thumbnails on Cloudflare and ads on Sume, set fit: crop and fit: cover and you will get the same framing from both.
Sources
Related posts
More in Media tools
- Cloudinary ar_9:16,c_fill,g_auto vs Sume per-shot crop fractions
Cloudinary reframes video with ar_9:16,c_fill,g_auto. On Sume you pick crop x per shot (left, centre, right), with a runnable Python script.
- Cost of subtitling 20 Shorts in 5 languages: 100 caption jobs
20 Shorts x 5 languages is 100 Video captions jobs. Price it at $0.20 each, plan the idempotency keys, and know what a failed job costs.
- Flow edit modes Insert, Remove, Camera: the prompt wording for Sume
Flow's Omni edit offers Insert, Remove and Camera modes; Sume has one video_url edit and a prompt. Prompt templates for each mode and what to check.
- Flow keeps 3 edit turns and History; build the audit trail on Sume
Flow's Omni edits hold context for up to 3 conversational turns and list every version in History. On Sume, each edit is a job id you can log and replay.
Written by Sume